摘要
为实现对V型坡口焊缝的精确检测,提出了一种基于结构光视觉传感器的弧焊机器人检测系统方法。该方法实时采集线结构光焊缝图像,采用中值滤波和最大类间方差法对图像进行去噪和阈值分割后,运用细化及斜率分析法提取结构光条纹中心直线,最后结合机器人手眼及光平面标定参数获取焊缝检测特征直线方程和特征点坐标。通过V型坡口焊缝检测实验,验证了该系统的特征点坐标检测相对误差在0.5%以内,具有较好的检测精度,满足工业实际现场的要求。
In order to realize the accurate detection of V-groove weld seam, a new method of detection for arc welding robot based on structured light vision sensor is proposed. structured light vision sensor is used to acquisite the real-time seam image. Median filtering and maximum between-cluster variance algorithm are used for noise reduction and threshold segmentation, after that, thinning and slope analysis method is used to extract the center of structured light stripe line. Finally, combing with the robot hand-eye and structured light plane calibration parameters, the information of weld seam detection is obtained. The V-shape groove weld detection experiments show that the absolute error of feature points is within 0.5%, so visual system has favorable detection precision and meets actual requirements.
引文
[1] 范明洋, 嵇保健, 洪磊. 基于线结构光的曲线焊缝焊接技术[J]. 组合机床与自动化加工技术, 2017(9): 142-145.
[2] 郭根, 习俊通, 陈晓波. 基于双目结构光的大型薄壁构件在线壁厚检测技术[J]. 组合机床与自动化加工技术, 2018(3): 62-70.
[3] 王邦国. 薄板激光拼焊视觉检测系统标定方法研究[J]. 机械设计与制造, 2015(8): 117-120.
[4] L Nele, E Sarno, A Keshari. An image acqui- sition system for real-time seam tracking[J]. Int. J. Adv. Manuf. Technol. 2013, 66 (9-12): 2099- 2110.
[5] M Dinham, G Fang. Autonomous weld seam identification and localization using eye-in-hand stereo vision for robotic arc welding[J]. Robot. Comput.–Integr. Manuf,2013, 29(5):288-301.
[6] Z Ye, G Fang, S B Chen,et al, Passive vision based seam tracking system for pulse-MAG welding[J]. Adv. Manuf. Technol, 2013,67(9-12): 1987-1996.
[7] 徐德, 谭民, 李原. 机器人视觉测量与控制[M].北京: 国防工业出版社, 2016.
[8] 李金伦, 崔少辉. 汪明. 基于改进中值滤波和提升小波变换的阈值去噪方法研究[J]. 应用光学,2014, 35(5):817-822.
[9] 张婧茹, 莫锦秋, 曹家勇. 一种鲁棒的室外金属标记物图像分割算法[J]. 机械设计与研究, 2016, 32(3): 31-37.
[10] 张万江, 许敏. 基于结构光视觉的激光拼焊焊缝质量检测方法研究[J]. 机械设计与制造,2011(3): 156-158.